completecase

Completecase refers to complete-case analysis, a simple method for handling missing data in statistical analyses by restricting the dataset to observations that have no missing values for the variables being analyzed. It is sometimes called listwise deletion because any case (row) with a missing value is removed from the analysis.

In practice, the approach involves identifying observations with complete data for all variables of interest and discarding all others before proceeding with standard statistical methods. It is widely used for its simplicity and interpretability, particularly in regression models, t-tests, and other conventional analyses.

Statistical properties of complete-case analysis depend on the mechanism causing missing data. If data are missing completely at random (MCAR), complete-case analysis yields unbiased parameter estimates, albeit with reduced efficiency due to a smaller sample size. If data are missing at random (MAR) or not at random (MNAR), estimates can be biased, and the bias may depend on the nature of the missingness and the variables included in the analysis. The main drawback is loss of information and reduced statistical power, which can also affect the precision of estimates and the ability to detect effects.

Implementation varies by software. In R, the complete.cases function identifies complete observations, which can be used to subset a data frame. In Python's pandas library, dropna removes rows with missing values. Other statistical packages provide analogous listwise deletion options.

Alternatives to complete-case analysis include multiple imputation, maximum likelihood methods that handle missing data directly, and inverse probability weighting, which can reduce bias under MAR. See also missing data; MCAR, MAR, MNAR; imputation.